No Comment — Guild Wars 2 Forums

No Comment

bobsort.4097bobsort.4097 Member ✭✭✭
edited January 31, 2020 in Account & Technical Support

This keep happening since Christmas patch (winter something events).
At first it was just TP that wouldn't open and
Then using jackal become impossible and
then since Lunar thingy event I can't even do anything in game.
latency keep jumping between 350 to 3000 and sometimes even to 8000.

I'm sure there is no issue in my computer and my connection until amazon servers. whatever issue it is, happening inside amazon servers which I can't even trace.
Support tickets are no use either, 3 support request (2 in game) and one via web and none of them was responded or reported issues fixed.

I'm not trying to give constructive feedback or suggestion. I need a fix so I can play the game that I paid for it.

1- First image is my firewall list of connections between game and server.
2- other 3 are trace route of those connections. as you can see I have minimal latency between my computer and amazon servers. and after that all I get is timeout (some because that specific IP is blocking ping, but at least the last hop should be reachable. either anet locked ping on their server or I cannot reach them. either way I cannot help more than this. this is all the information I have for you.


  • Well, as you can see, you start having issues when you hit the Tier 1 Backbone, Telia.
    You can't ping Amazon servers; AWS doesn't allow it.
    You can try a VPN that takes a different route to the GW2 servers.

    Also, as indicated on the in-game Bug Report window, no response will be forthcoming.
    Every Support Ticket is responded to; if you did not receive the near-immediate automated reply, check your Spam/Junk folders.
    Do keep in mind, the CS Team allows 72 hours for a Team-member response to a ticket.

    Good luck.

  • bobsort.4097bobsort.4097 Member ✭✭✭

    Thanks for reply. ping latency of between 200 to 500 is usual for me and I don't have any issue over there. I have lost hope to have anything less than that long ago.

    Following picture is from this morning that I can login and play without issue. As you can see current ping is 260. that may be too much for players who used to play in pings like 10ms but for me it is enough. Since I can play now, I can come to conclusion that 1 support fixed the issue and second I don't need to include a ping screenshot, just including IP would be enough.

    Just a side note, I have few servers of my own that act like this once in a while, and our solution was to schedule restart every week. Last time we see such issues (disconnect, unable to connect, latency) was over a year ago. Since anet using windows servers too. I suggest disabling auto update on them and do it manually when they have tested the update in their test environment first. Microsoft has a great reputation on sending updates that breaks the production servers.

  • bobsort.4097bobsort.4097 Member ✭✭✭

    False alarm, still same issue. still latency spikes over the roof

  • Hannelore.8153Hannelore.8153 Member ✭✭✭✭
    edited February 1, 2020

    @bobsort.4097 said:
    False alarm, still same issue. still latency spikes over the roof

    The problem isn't AWS, its Telia, 177ms is too high of a ping to have on your first external hop and means something is wrong there.

    At one point you also get 165ms to a local IP (bad wireless, copper or coaxial SNR).

    Hannah | Daisuki[SUKI] Founder, Ehmry Bay, NA | 22 charas, 16k hours, 27k AP | ♀♥♀
    Mains Mariyuuna/Tempest & Alisha Kei/Druid(PvE), Terakura/Spellbreaker & Kitty Koume/Reaper(WvW)
    No need to be best, only good and kind.

  • bobsort.4097bobsort.4097 Member ✭✭✭

    As I mentioned in my second post. having ping between 200 to 500 is usual in SEA area. Also Telia is just one more hop to Amazon server. and as you can see from pings, amazon ping (hope 25,26) is still in acceptable range for me. Also I'm not on copper, coaxial or wireless. I'm on fiber connected to a gigabit lan router and with cat6 cable to gigabit card on my computer.

  • Healix.5819Healix.5819 Member ✭✭✭✭

    You potentially have some packet loss starting at hop 7 in two of the traces, but tracert isn't a good sample. The built-in option to use is pathping, but it has the flaw of stopping at the first hop that doesn't respond. Ideally you'd want to run PingPlotter in the background while you play to get a sample of the spike (use the IP of the map you're in).

    You can try using the -clientport option with either 80 or 443 in hopes your ISP treats it differently, otherwise you can only try a VPN:
    -clientport 80

  • bobsort.4097bobsort.4097 Member ✭✭✭

    I believe Hop 7 is the dns server which is normally don't respond to pings

    It is not an spike. it is complete loss of connection most of the time. And this is not the issue with jakal mount that if I jump through multiple ports, I get disconnected (normally second one)

    I'm not sure how pingplotter can help that ping cannot do, but here you go. again have to mention, 300~500ms latency for me which connecting from SEA is considered playable. but not anything above 500ms and I don't remember having any latency below 300.